INDEX

Explanations

instances of disruption or interruption

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 отношению

-0.63

rawDesc

-0.47

-0.46

Sta

-0.46

Sel

-0.46

 aank

-0.45

Extent

-0.44

copyOf

-0.43

ál

-0.43

 tài

-0.43

POSITIVE LOGITS

 disrupt

1.17

 disrupting

1.14

 disrupted

1.12

 disruption

1.10

 disruptions

1.06

 disturbance

1.01

rupts

0.98

 disruptive

0.97

 disturbances

0.94

 interrupting

0.94

Activations Density 0.243%

instances of disruption or interruption

No Comments

No Known Activations